Facial Expression Recognition Using Independent Component Features and Hidden Markov Model
Md. Zia Uddin, Tae‐Seong Kim · 2015
Facial expression recognition (FER) from video is an essential research area in the field of human-computer interfaces (HCIs). This chapter presents a new method to recognize several facial expressions from time sequential facial expression images. It proposes a novel approach for FER dealing with enhanced independent component analysis (EICA), Fisher linear discriminant analysis (FLDA), and hidden Markov models (HMMs). The chapter also presents the proposed system methodology from video preprocessing to expression training and recognition using HMMs. It represents the experimental results obtained using the proposed approach on the Cohn-Kanade expression database. The chapter concludes that the holistic face features can be used for the HMM method rather than usage of action units for HMM to recognize the partial face activities.